[S-CG45] Science of slow-to-fast earthquakes

[SCG45-14] Toward Improved Urban Earthquake Monitoring through Deep-learning-based Noise Suppression

★Invited Papers

*Lei YANG^1,2,3, Xin Liu⁴, Liang Zhao², Weiqiang Zhu⁵, Gregory C. Beroza³ (1.Earthquake Research Institute, the University of Tokyo, 2.Institute of Geology and Geophysics, Chinese Academy of Sciences, 3.Department of Geophysics, Stanford University, 4.University of Hong Kong, 5.Department of Earth & Planetary Science, University of California)

Keywords:earthquake monitoring in urban settings, seismic denoising, deep learning, dense nodal array

Earthquake risk is highest in urban settings, owing to the population density and the presence of extensive and vulnerable infrastructure. Ideally, the intensive earthquake monitoring efforts in urban areas would be used to characterize the fault systems that pose the most immediate and direct threats to cities. However, the same factors—population and infrastructure—that cause risk exposure to be high also make earthquake monitoring difficult to carry out. This is due to the various types of seismic noise generated in cities and the logistical difficulties of instrument deployment.
For an improved earthquake monitoring in urban areas, we develop a deep-learning-based denoising algorithm, UrbanDenoiser, to filter out urban seismological noise. UrbanDenoiser strongly suppresses noise relative to the signals, because it was trained using waveform datasets containing rich noise sources from the urban Long Beach dense array and high signal-to-noise ratio (SNR) earthquake signals from the rural San Jacinto dense array. Application to urban seismic data shows that UrbanDenoiser can recover signals with little phase distortion in body waves, and the robust seismic denoising allows us to work on the entire day’s data, not just during the night when anthropogenic noise is lower, which doubles the utility of existing data. We also apply UrbanDenoiser to a regional seismic network for the La Habra earthquake sequence in the urban area. The denoising method can recover signals at an SNR level down to ~0 dB, and leads to an increased detection rate amounting to more than 4.5 times the number of detections in the Southern California Seismic Network catalog. Earthquake location using the denoised Long Beach data does not support the previous report of mantle seismicity beneath Los Angeles, but suggests a fault model featuring shallow creep, intermediate locking, and localized stress concentration at the base of the seismogenic zone.